#EBM #Statistics

Intro

Data

400

Errors

  1. Systematic = bias เพี้ยนอย่างมี direction แก้ด้วย study design คุณภาพ
  2. Random = chance เพี้ยนมั่ว (unpredictable) แก้ด้วย statistic ที่ดี

400

Randomization

3 Keys-> ชนิดไหน, ปกปิดยังไง, balance ไหม

Study designs

Biases

Types Solutions
selection balanced randomization

Details

Biostats

400

400
400

400

400
400

Bio-informations

Keys -> Scopus 1960 > Pubmed 1945 > Medline
- Pubmed search[options] L-R
- Scopus options["search"] OR-AND-NOT

Pubmed

Keys -> ดู search detail ทุกครั้ง, MeshTerm, ATM

Automatic term mapping(ATM)

ทำให้ค้นหาง่าย easy keyword เพราะจะแมพเข้า meshterm อัตโนมัติเลย

Meshterm (Medical subject heading)

ช่วยให้หาได้กว้างขึ้น โรค/ภาวะเดียวกันพิมพ์คำ variation ไหนก็เจอ สะดวกดีนะ

ห้องสมุด

Scholar Scopus Pubmed
med and non-med med med
ชินกับคำค้น Meshterm ก็มี, tagging ทำให้หาง่ายกว่าในด้าน filter, ภาษามากกว่า update กว่า pubmed

Analysis

Outcome Test
categorical linear
continuous logistic
time to event cox proportional hazard
incidence rate poisson

Diagnostic

Prognosis

400

RCT

Keys -> intervention ทำ NI ก่อนไป superior (Heirachy testing)

Systematic review and metaanalysis

Economic

Ethics

Spot dx

400

MEQ

  1. สรุป paper นี้ น่าเชื่อถือไหม ?
  2. สรุป paper นี้ เอาไป apply ได้ไหม ?